In a device with both a speech user interface (SUI) and a Graphical User Interface (GUI), the user is typically faced with learning separate command sets for each of the SUI and GUI. This increases the difficulty of accurately giving commands, especially for the SUI. In addition, speech commands are difficult to remember since they are transient and there is no persistent display of them to help the user remember. Existing systems fail to provide a mechanism to unify the command sets of the GUI and SUI so that the user need only learn a single instance of many of the commands to use in both interfaces.
U.S. Pat. No. 6,075,534 and others describe methods for displaying a dynamically changing menu to control and provide feedback on the operation of the speech recognizer in a GUI. However, no reference can be found that ensures that the GUI menus and dialog box elements are constructed in such a way that by selecting them the user essentially builds the corresponding speech command for an application.